Loctite, Brass & Steel

[petertha]

I tried the same Loctite retainer on the carbon steel shim on my 1018 scrap shaft. Same result, peeled off just as easily as the brass. You can see the green stripe within the red felt pen lines where its entirely cured but just no gription.

Then I tried medium viscosity CA glue (between the blue lines). It stuck better, but not anything like what I would call a permanent bond. It would probably stay put if you were sliding a loose fit bearing over it but I'm not sure how much shear stress it would take before failing

Very strange. I don't know what to say about this anymore. I know CA prefers a bit of porosity. But makes me wonder how bearings can stay so firmly stuck to a shaft with normal Loctite retainer where you sometimes have to beat them or torch heat to remove. Maybe its purely better strength in shear (like pushing a bearing off along a shaft or machining a part bonded to an arbor face) vs tensile strength (peeling back a thin stock). Interestingly the 'glue' is always fully cured to the shaft but not to the shim stock, I need to talk to a Loctite engineer, this will follow me to my grave LOL!

 

[petertha]

I never really mentioned what this was all about but involved helping a fellow restore a 1940s Walker Turner 900 drill press. The shaft was a bit undersized to a 0.625" bearing, about 0.001" annular gap. He already had new bearings but we went through the various permutations of a replacing them with a next larger metric (17mm ID) bearing & machining a sleeve. But it was more complicated than that. The OD had to also fit an IMP quill recess. But the bigger challenge was the specific sequence - install lower bearing on spindle, then inner race spacer tube, then quill which fits bearing OD, then upper bearing, then a threaded backlash fitting, then...

Anyways I was able to carefully sneak the 0.001" shim stock inside the gap between bearing using my fingernails & work my way up the assembly step by step. The shim is so thin its just like weak foil that is easy to crumple if you look at it wrong. After some careful truing of some chowdered surfaces the reward was a slop free, nicely rotating spindle with about 0.001" runout on the tapered arbor which is awesome for a vintage machine resto project. He did a great job bringing it back to this state so I hope it brings many happy years of accurate drilling.

 

[petertha]

Hope he doesn't mind me posting a pic of his work in progress. Cool looking vintage iron, eh?

 

[hman]

But makes me wonder how bearings can stay so firmly stuck to a shaft with normal Loctite retainer where you sometimes have to beat them or torch heat to remove. Maybe its purely better strength in shear (like pushing a bearing off along a shaft or machining a part bonded to an arbor face) vs tensile strength (peeling back a thin stock).

Think of it this way ... when a bearing is "glued" to a shaft, and you're dealing with shear, you have an area effect - the circumference of the shaft times the height of the bearing. When peeling a shim off a shaft, you're concentrating all your force into a tiny area - the width of the shim times some tiny width of contact area where the shim is being lifted off - maybe a hundredth of an inch?

 

[petertha]

Yes, that's how I'm visualizing it now. We never peel a bearing race off like the shim. Whether it is pushed along the shaft or rotated relative to the shaft, the 'adhesive' is resisting in pure shear modes with all its surface area. Not tension focused to the teeny leading edge area when peeling off a band. I want to machine a simple collar with a known (thin) annular gap and try & pure torque or axial displacement in this manner.

 

[Flyinfool]

I am glad you managed to get it done. Sometimes perseverance pays off.

In the Loctite literature they refer to active vs inactive surfaces. Loctite only cures with an active surface. Steel is an active surface. Black oxide (or many other finishes) will make the surface inactive. brass is an inactive surface. At least one of the surfaces must be active to be able to activate the loctite. Primer helps but is still not as strong as steel to steel. Ideally you want both surfaces to be active, that is why ALL of there strength charts are based on steel to steel.

As mentioned above, Loctite has very low peel strength, it was not designed for that. If you were to Loctite your shim to the bearing and the shaft, in use it would hold fine since there is no peel load, just the shear load it was designed for.

 

[petertha]

A few more pics to log. I tried regular JB Weld with both brass & steel. It seemed to hold a bit better than Loctite or CA. I was able to turn it down to basically nothing. So might be a cheesy but still viable method to 'build up' worn or mismatched shaft thickness to a bearing ID.

But again, once I lifted up an edge I could peel back the shim without a lot of effort. So I think our observations are right. The 'peeling back' action is putting very equivalent tensile stress localized on a teeny pencil line width at the leading edge line of adhesive. Once it lifts (fails), it perpetuates along the joint. I don't think its related to the material itself at all, but purely to this effect. A rigid bearing or collar could never be peeled back like this which is why we dont see it in real life. It needs to be pushed off axially, therefore the entire bonded surface area is resisting and in shear mode vs tensile, which is more favorable to the adhesive. At least that's my armchair engineering notion.

 

[pontiac428]

I'm wondering if the brass shim sheet is too smooth for good adhesion. Maybe a bead blast profile or a good roughing up with some coarse emery cloth will help it stick?

 

[petertha]

I don't have blasting gear. I agree, a mat finish would probably be the best. But but all the shim material was scuffed at two 90-deg orientations with about 220g paper & wiped clean with acetone. So it was shiny new material & should have at pretty good tooth for the adhesive. It was pretty much the exact same story for brass, steel & aluminum on the steel mandrel.

I'm going to machine a simple close fitting collar of some thickness, say 1/8" or so & try the same JB weld. I'm guessing that it should hold up fine to axial pushing load. But that's very different than peeling it back mode like with the thin shim stock.

 

[Flyinfool]

I would still like to see your tests done after soft soldering the brass shim in place.